[RHEL7,COMMIT] sched: really don't call calc_load_ve() under jiffies_lock

Submitted by Konstantin Khorenko on April 1, 2020, 5:37 p.m.

Details

Message ID 202004011737.031Hb2hH011365@finist-ce7.sw.ru
State New
Series "sched: really don't call calc_load_ve() under jiffies_lock"
Headers show

Commit Message

Konstantin Khorenko April 1, 2020, 5:37 p.m.
The commit is pushed to "branch-rh7-3.10.0-1062.12.1.vz7.145.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git
after rh7-3.10.0-1062.12.1.vz7.145.5
------>
commit 0e259098427c22420260c03feae705c8c1a90cc5
Author: Konstantin Khorenko <khorenko@virtuozzo.com>
Date:   Tue Mar 31 16:01:18 2020 +0300

    sched: really don't call calc_load_ve() under jiffies_lock
    
    Previously we've done all preparation work for calc_load_ve() not being
    executed under jiffies_lock, and thus not called from
    calc_global_load(), but forgot to drop the call in calc_global_load().
    So now we still call expensive calc_load_ve() under the jiffies_lock and
    get NMI.
    
    Fix that.
    
    Fixes:19bc294a5691d ("sched: Call calc_load_ve() out of jiffies_lock")
    
    https://jira.sw.ru/browse/PSBM-102573
    
    Signed-off-by: Konstantin Khorenko <khorenko@virtuozzo.com>
    Signed-off-by: Valeriy Vdovin <valeriy.vdovin@virtuozzo.com>
---
 kernel/sched/core.c | 2 --
 1 file changed, 2 deletions(-)

Patch hide | download patch | download mbox

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index c8e1aaaeefcbb..542fb96f4dd55 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3258,8 +3258,6 @@  bool calc_global_load(unsigned long ticks)
 
 	calc_load_update += LOAD_FREQ;
 
-	calc_load_ve();
-
 	/*
 	 * In case we idled for multiple LOAD_FREQ intervals, catch up in bulk.
 	 */